well but isn't that the point? your mempool doesn't matter because I can send it directly to the miner instead?
Login to reply
Replies (1)
Not sure which point you are trying to make but due to the OOB delivery always being possible and carrier size being a mempool issue, the knots crowd is fighting over mempool and explicitly claim ignorance about what happens in blocks. I think that is misguided. They should pivot to focusing on the public perception of what is invited to be done on the Bitcoin network and not cry over some low-res CSAM briefly being in somebody's node without their knowledge.
By deprecating carrier size settings, Core signaled that protocols can do with OP_RETURN whatever they want. They claim that serious protocols **sadly** wouldn't create 100kB OP_RETURNs because they would miss out on the SegWit discount.
Our worry shouldn't be serious protocols though. Our worry is malicious actors claiming to be serious. They can now spam OP_RETURN with "continuous bites" of nasty pictures that you would "not need specialized software to see". I put that in quotes as there already exists software to extract images from corrupted disk dumps for example. So yes, there would be software that could find these images but I find that a moot point as it's totally not something anybody would do accidentally. Is the mempool serialized to disk in unscrambled form? I guess I could find out but so I could find out how to re-assemble pictures from more obscure protocols so the whole distinction between scrambled in blocks but continuous in mempool is moot.